Gibbs Free Energy and Phase Stability in Functional Materials: Free-Energy Curves, Common Tangents, and Phase Separation
Gibbs Free Energy and Phase Stability in Functional Materials: Free-Energy Curves, Common Tangents, and Phase Separation
Why does a ceramic form a single solid solution at one temperature but separate into two phases at another? Why can a dopant have a limited solubility in an oxide, even when the nominal composition is chemically reasonable? Why can a material remain apparently single-phase after synthesis even though a two-phase assemblage has a lower equilibrium free energy?
These questions are addressed by Gibbs free-energy analysis. At fixed temperature and pressure, the stable equilibrium state is the state that minimizes the total Gibbs free energy subject to conservation of the overall composition. The shape of the free-energy–composition curve determines whether a homogeneous phase is stable, metastable, or unstable, and whether two phases can coexist.
This article develops the graphical and quantitative tools used to analyse phase stability in functional materials. It focuses on free-energy curves, common-tangent construction, miscibility gaps, spinodal decomposition, and the distinction between thermodynamic stability and kinetic persistence.
1. The equilibrium criterion for phase stability
For a closed system at constant temperature and pressure, equilibrium corresponds to a minimum in Gibbs free energy:
[ G=H-TS ]
A phase is not stable simply because its crystal structure is familiar or because it can be synthesized. Its stability must be assessed relative to all competing phases and phase assemblages that satisfy the same overall composition and external conditions.
For a system with several phases,
[ G_{}=_{} n^G_m^ ]
where (n^) is the amount of phase (), and (G_m^) is its molar Gibbs free energy. The equilibrium phase assemblage minimizes (G_{}), while conserving the total amount of every component.
This distinction is important: the free energy of a homogeneous phase at the overall composition is not necessarily the lowest free energy attainable by the system. A mixture of two phases with different compositions may have a lower total Gibbs free energy.
For the thermodynamic potentials and their natural variables, see Thermodynamic Potentials in Materials Science.
2. Gibbs free energy as a function of composition
Consider a binary system containing components A and B. Define the mole fraction of B as
[ x=x_B,x_A=1-x ]
At fixed temperature and pressure, the molar Gibbs free energy of a homogeneous phase can be represented as a function of composition:
[ G_m=G_m(x) ]
A free-energy–composition diagram plots (G_m) vertically against (x) horizontally. The curvature and relative position of the free-energy curves contain information about phase stability.
A minimum in a curve is not, by itself, sufficient to establish global stability. The relevant comparison is between the free energy of the homogeneous state and the free energy of every allowed phase-separated state.
For a single homogeneous phase at composition (x_0), its molar free energy is simply
[ G_{}=G_m(x_0) ]
If the system separates into phases () and (), with compositions (x_) and (x_), the overall composition must satisfy the lever-rule balance:
[ x_0=f_x_+f_x_ ]
where
[ f_+f_ ]
and (f_) and (f_) are the phase fractions on the same molar basis used to define (x).
The molar free energy of the two-phase mixture is
[ G_{} =f_G_m^(x_) +f_G_m^(x_) ]
If (G_{}<G_{}), phase separation lowers the Gibbs free energy and is thermodynamically favourable.
3. The common-tangent construction
The common-tangent construction is a graphical method for identifying the compositions of two coexisting phases in a binary system at fixed temperature and pressure.
Draw a straight line that is tangent to the free-energy curve at two compositions, (x_) and (x_). The two tangent points define the equilibrium compositions of the coexisting phases, provided the phases belong to the free-energy branches being compared and the construction represents the global lower free-energy envelope.
Mathematically, the tangent slopes must be equal:
[ .|{x} = .|{x} =m ]
The intercepts must also be equal. Equivalently,
[ G_m^(x_)-mx_ = G_m^(x_)-mx_ ]
Together, these conditions mean that the partial molar Gibbs energies of both components are equal between the two phases:
[ _A=_A ]
[ _B=_B ]
These chemical-potential equalities are the fundamental equilibrium conditions. The common tangent is their graphical expression for a binary system.
For the relation between chemical potential, activity, and equilibrium, see Chemical Potential and Activity in Solid Materials.
3.1 Why a straight line matters
The line joining two candidate phase compositions represents the molar free energy of a phase mixture at intermediate overall compositions. If the line lies below the homogeneous free-energy curve, the mixture has a lower free energy than the single phase.
At equilibrium, the common tangent gives the lowest possible linear combination of phase free energies for the specified overall composition. In geometric terms, the stable equilibrium free energy is the lower convex envelope of the available free-energy branches.
A curve can therefore lie above the common tangent over a composition interval even if it is locally smooth and corresponds to a physically realizable homogeneous solid solution. That interval is unstable against decomposition into the two tangent compositions.
4. The lever rule: calculating phase fractions
Once the equilibrium compositions (x_) and (x_) have been identified, the phase fractions follow from mass balance.
For an overall composition (x_0) between the two coexistence compositions,
[ f_= ]
[ f_= ]
These expressions satisfy
[ f_+f_ ]
Worked example
Suppose a binary oxide system at a selected temperature has two coexisting solid-solution compositions:
[ x_,x_ ]
A specimen has an overall composition
[ x_0=0.35 ]
The equilibrium phase fractions are
[ f_==0.75 ]
[ f_==0.25 ]
Thus, on the chosen molar basis, the equilibrium assemblage contains 75% phase () and 25% phase ().
The lever rule calculates phase fractions only after the equilibrium phase compositions are known. It does not determine the coexistence compositions; those come from the free-energy functions or experimental phase-equilibrium data.
5. Ideal and regular-solution models
A simple model helps illustrate how enthalpy and entropy compete to determine phase stability. For an ideal binary solution,
[ G_m=x_A G_A^+x_B G_B^ +RT(x_Ax_A+x_Bx_B) ]
where (x_A=1-x) and (x_B=x).
The logarithmic terms represent ideal configurational entropy. Their contribution lowers the free energy at intermediate compositions and favours mixing.
Real solid solutions can exhibit non-ideal interactions. A regular-solution model introduces an interaction parameter ():
[ G_m=x_A G_A^+x_B G_B^ +RT(x_Ax_A+x_Bx_B) +x_Ax_B ]
The excess free-energy term is
[ G_m^{}=x_Ax_B ]
For a symmetric regular solution:
- (>0) represents an energetic penalty for unlike-neighbour mixing and can promote phase separation.
- () gives the ideal-solution model.
- (<0) favours unlike-neighbour interactions and tends to stabilize mixing.
The regular-solution model is an illustrative approximation. Real oxides may require composition-dependent interaction parameters, sublattice models, defect models, or explicit treatment of ordering and multiple crystallographic sites.
The entropy contribution and its microscopic meaning are discussed in Entropy in Materials Science.
6. Binodal and spinodal limits
A miscibility gap is a composition range in which a homogeneous solution is not the equilibrium state and separates into two phases of different compositions.
Two boundaries are useful when describing the stability of a homogeneous solution: the binodal and the spinodal.
6.1 Binodal: the coexistence boundary
The binodal, also called the coexistence curve, is defined by the two compositions connected by the common tangent. At the binodal compositions, the two phases have equal chemical potentials for each component.
Inside the miscibility gap, the equilibrium state is a two-phase mixture. Outside it, a homogeneous phase may be the equilibrium state, subject to comparison with all other possible phases.
The binodal is a global equilibrium boundary: it compares the homogeneous phase with phase-separated alternatives.
6.2 Spinodal: the limit of local stability
For a homogeneous binary solution described by (G_m(x)), local stability against infinitesimal composition fluctuations requires positive curvature:
[ >0 ]
The spinodal boundary is defined by
[ =0 ]
Inside the spinodal region,
[ <0 ]
and infinitesimal composition fluctuations lower the free energy. The homogeneous state is locally unstable and can decompose spontaneously by spinodal decomposition, without requiring nucleation of a distinct phase.
Between the binodal and spinodal boundaries, the homogeneous phase is metastable. A finite fluctuation or nucleus is required to initiate phase separation, because small fluctuations increase the free energy.
The distinction is therefore:
| Region | Thermodynamic status of homogeneous phase | Expected decomposition pathway |
|---|---|---|
| Outside binodal | Stable against the considered demixing process | No equilibrium driving force for demixing |
| Between binodal and spinodal | Metastable | Nucleation and growth |
| Inside spinodal | Locally unstable | Spinodal decomposition |
The precise boundaries depend on temperature and the free-energy model. In multicomponent systems, stability is determined by the full curvature matrix (Hessian), not by a single second derivative.
7. Critical point in a symmetric regular solution
For the symmetric regular-solution model, the free energy can be written, apart from the linear end-member reference terms, as
[ G_m=RT[xx+(1-x)(1-x)]+x(1-x) ]
The critical point occurs where the second and third derivatives with respect to composition vanish:
[ =0, =0 ]
For this symmetric model, the critical composition is
[ x_c= ]
and the critical temperature is
[ T_c= ]
for positive, temperature-independent ().
Above (T_c), the ideal entropy contribution is sufficiently strong to stabilize a homogeneous solution for this model. Below (T_c), a miscibility gap appears.
This result is model-specific. Asymmetric interactions, multiple sublattices, ordering transitions, or additional phases can shift the critical composition or produce more complex phase diagrams.
8. Applying free-energy analysis to functional oxides
The same principles apply to oxide ceramics, but the composition variable may represent cation substitution, oxygen deficiency, proton content, or a coupled set of defect populations.
8.1 Cation solid solutions and dopant solubility
Consider a perovskite oxide in which a dopant substitutes on the A or B sublattice. A single-phase solid solution is stable only if its Gibbs free energy is lower than that of all competing phase assemblages at the same overall composition.
A nominal dopant concentration does not guarantee that all dopant atoms enter the target lattice. If a secondary oxide or another perovskite has a lower combined free energy, the material may form a multiphase assemblage.
This is why phase-purity assessment requires more than checking whether the nominal composition lies within a broad structural family. Synthesis temperature, oxygen chemical potential, cation volatility, and competing phases all affect the relevant free-energy landscape.
8.2 Oxygen nonstoichiometry
For a non-stoichiometric oxide such as (ABO_{3-}), oxygen content is a compositional variable. The equilibrium value of () depends on temperature and oxygen chemical potential.
A free-energy model must account for the energetic cost of forming defects and the entropy associated with their distribution. The equilibrium oxygen content is obtained by minimizing the appropriate thermodynamic potential under the imposed gas conditions and charge-neutrality constraints.
Experimental measurements of ((T,p_{})) provide essential constraints for validating such models. See Measuring Oxygen Nonstoichiometry in Functional Oxides.
8.3 Proton incorporation and hydration
In proton-conducting oxides, hydration changes the concentrations of oxygen vacancies and protonic defects. A free-energy description must include the chemical potentials of the solid components and water vapour, as well as the configurational and other relevant entropy contributions.
The equilibrium proton content is not determined by the hydration reaction alone. It also depends on the available vacancy population, dopant concentration, temperature, water chemical potential, and competing defect states.
See Proton-Conducting Oxides for the defect-chemistry and transport context.
9. Thermodynamic stability versus kinetic persistence
A phase can persist experimentally even when it is not the global equilibrium state. This does not invalidate the free-energy analysis; it means that the system has not reached equilibrium on the available timescale.
Several kinetic factors can delay transformation:
- Slow cation diffusion in a dense ceramic.
- Limited oxygen exchange or slow defect redistribution.
- Nucleation barriers for a competing phase.
- Grain-boundary or interface constraints.
- Processing-induced strain and residual stress.
- Cooling rates that preserve a high-temperature phase assemblage.
A metastable phase may therefore be retained after quenching or low-temperature processing. Conversely, an apparently single-phase X-ray diffraction pattern does not prove thermodynamic stability: a small fraction of a secondary phase may be below the detection limit, poorly crystalline, or structurally similar to the main phase.
It is useful to distinguish three statements:
- Thermodynamically stable: the state has the lowest Gibbs free energy under the specified conditions.
- Metastable: the state is a local minimum but not the global minimum.
- Kinetically retained: the state persists because transformation is too slow or inhibited, whether or not it is a local minimum in a simplified model.
10. Practical workflow for evaluating phase stability
A practical analysis can be organized as follows:
- Define the system. Specify the components, allowed phases, temperature, pressure, and relevant gas reservoirs.
- Choose the composition variables. Identify whether composition refers to cation fractions, oxygen deficiency, proton content, or sublattice occupancies.
- Construct free-energy models. Include end-member contributions, configurational entropy, non-ideal interactions, and relevant defect or ordering terms.
- Compare competing phases. Use consistent reference states and thermodynamic conventions.
- Determine coexistence conditions. Apply equality of component chemical potentials or a common-tangent construction for a binary system.
- Calculate phase fractions. Apply mass balance through the lever rule.
- Assess local stability. Examine the curvature of the free-energy function, or the Hessian for multicomponent systems.
- Consider kinetics. Determine whether diffusion, nucleation, and gas exchange are fast enough for equilibrium to be approached.
- Validate experimentally. Combine diffraction, composition analysis, thermal analysis, and controlled-atmosphere measurements as appropriate.
For quantitative databases and multicomponent phase-equilibrium calculations, CALPHAD methods provide a systematic framework. Their predictions remain dependent on the quality, range, and consistency of the thermodynamic descriptions used.
11. Common mistakes
- Comparing only the free energy of one homogeneous phase. A two-phase mixture may have a lower total free energy.
- Using a tangent at only one composition. Two-phase coexistence requires a common tangent touching both relevant free-energy branches.
- Confusing the binodal and spinodal. The binodal is a global coexistence boundary; the spinodal marks loss of local stability.
- Treating a metastable phase as the equilibrium phase. Experimental persistence can reflect kinetic limitations.
- Applying the lever rule before determining coexistence compositions. Phase fractions require the binodal compositions as input.
- Ignoring the composition basis. Phase fractions and composition balances must use a consistent molar or mass basis.
- Assuming a regular-solution model is universally quantitative. It is a simplified model and may not describe ordering, defect association, or multiple sublattices adequately.
12. Summary
At constant temperature and pressure, phase stability is determined by minimizing the total Gibbs free energy while conserving the overall composition. A homogeneous phase can be less stable than a mixture of two phases, even when its own free-energy curve is smooth.
The common-tangent construction identifies the compositions of coexisting phases, and the lever rule gives their fractions. The binodal defines the coexistence boundary, while the spinodal marks the loss of local stability. Between these boundaries, a homogeneous material is metastable and generally requires nucleation to decompose.
For functional oxides, this framework helps interpret dopant solubility, secondary-phase formation, oxygen nonstoichiometry, hydration, and the difference between equilibrium stability and phases retained by processing.
The central practice is to compare complete free-energy alternatives under clearly specified conditions—not to infer stability from nominal composition, crystal structure, or experimental persistence alone.
Further reading
- D. R. Gaskell and D. E. Laughlin, Introduction to the Thermodynamics of Materials, 6th ed., CRC Press.
- H. B. Callen, Thermodynamics and an Introduction to Thermostatistics, 2nd ed., Wiley.
- A. R. West, Solid State Chemistry and its Applications, 2nd ed., Wiley.
- J. W. Cahn and J. E. Hilliard, “Free Energy of a Nonuniform System. I. Interfacial Free Energy,” The Journal of Chemical Physics 28, 258–267 (1958).